经验模态分解及其在降噪方面的应用

Empirical Mode Decomposition and Its Application on Signal Denoising

本文详细介绍了经验模态分解（EMD）方法，描述了EMD算法实现步骤；通过EMD分解，任何信号序列部町分解为一系列不同尺度的固有模态函数（IMF），这种分解方法是从信号本身的尺度特征出发对信号进行分解，具有良好的自适应性，非常适合对非线性非平稳信号进行分析，并列举实例证明了其有效性。同时，提出了一种基于EMD的小波阈值降噪方法，该斤法很大程度上克服了直接小波阈值降噪的一些缺陷，仿真数据处理证明了该方法的有效性。

This paper introduces the ＇empirical mode decomposition＇ of EMD. With EMD any complicated signal can be decomposed into a finite method（EMD）,list out the basic steps and often small number of ＇intrinsic mode function＇ （IMF）.Since this decomposition method is adaptive, and the decomposition is based on the local characteristic time scale of the data, it is applicable to nonlinear and non-stationary processes. Example is given to demonstrate the advance and efficient of the method. Meanwhile, a new wavelet threshold de-noising method based on EMD is proposed. This method avoids some deficiencies of the traditional de-nosing process. The simulation results show this method has advantage over the traditional wavelet threshold de-noising method.